Yeah, seen plenty of poorly calibrated risk matrices, but they are generally better than nothing. Also, for an organisation with multiple facilities, or using

Yeah, seen plenty of poorly calibrated risk matrices, but they are generally better than nothing. Also, for an organisation with multiple facilities, or using consultants or external resources it is the most effective practical way to convey the owners/boards risk appetite and willingness to spend money to avoid unwanted consequences. The text is long winded to the point of being obtuse and basic theme says that you need to understand the underlying mechanisms to effectively design one or apply use. No secrets there. I have just been thru an exercise where the clients consultant decided to instead generate some fault trees, in excel. 15000 formulas in each spreadsheet, 5000 indirects. Numerous errors, many of them of a fairly outcome sensitive nature. Now what do you think is the worse the, matrix or error riddled complexity with methodology unaccessible to most people? No easy answers and the point the article doesn't seem to make is you need to keep going back to your risk assessment and matrix calibration with acquired learnings, it should be an iterative process of refinement.